Free4D: Tuning-free 4D Scene Generation with Spatial-Temporal Consistency

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Liu, Tianqi, Huang, Zihao, Chen, Zhaoxi, Wang, Guangcong, Hu, Shoukang, Shen, Liao, Sun, Huiqiang, Cao, Zhiguo, Li, Wei, Liu, Ziwei
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866917968574676992
author Liu, Tianqi
Huang, Zihao
Chen, Zhaoxi
Wang, Guangcong
Hu, Shoukang
Shen, Liao
Sun, Huiqiang
Cao, Zhiguo
Li, Wei
Liu, Ziwei
author_facet Liu, Tianqi
Huang, Zihao
Chen, Zhaoxi
Wang, Guangcong
Hu, Shoukang
Shen, Liao
Sun, Huiqiang
Cao, Zhiguo
Li, Wei
Liu, Ziwei
contents We present Free4D, a novel tuning-free framework for 4D scene generation from a single image. Existing methods either focus on object-level generation, making scene-level generation infeasible, or rely on large-scale multi-view video datasets for expensive training, with limited generalization ability due to the scarcity of 4D scene data. In contrast, our key insight is to distill pre-trained foundation models for consistent 4D scene representation, which offers promising advantages such as efficiency and generalizability. 1) To achieve this, we first animate the input image using image-to-video diffusion models followed by 4D geometric structure initialization. 2) To turn this coarse structure into spatial-temporal consistent multiview videos, we design an adaptive guidance mechanism with a point-guided denoising strategy for spatial consistency and a novel latent replacement strategy for temporal coherence. 3) To lift these generated observations into consistent 4D representation, we propose a modulation-based refinement to mitigate inconsistencies while fully leveraging the generated information. The resulting 4D representation enables real-time, controllable rendering, marking a significant advancement in single-image-based 4D scene generation.
format Preprint
id arxiv_https___arxiv_org_abs_2503_20785
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Free4D: Tuning-free 4D Scene Generation with Spatial-Temporal Consistency
Liu, Tianqi
Huang, Zihao
Chen, Zhaoxi
Wang, Guangcong
Hu, Shoukang
Shen, Liao
Sun, Huiqiang
Cao, Zhiguo
Li, Wei
Liu, Ziwei
Computer Vision and Pattern Recognition
We present Free4D, a novel tuning-free framework for 4D scene generation from a single image. Existing methods either focus on object-level generation, making scene-level generation infeasible, or rely on large-scale multi-view video datasets for expensive training, with limited generalization ability due to the scarcity of 4D scene data. In contrast, our key insight is to distill pre-trained foundation models for consistent 4D scene representation, which offers promising advantages such as efficiency and generalizability. 1) To achieve this, we first animate the input image using image-to-video diffusion models followed by 4D geometric structure initialization. 2) To turn this coarse structure into spatial-temporal consistent multiview videos, we design an adaptive guidance mechanism with a point-guided denoising strategy for spatial consistency and a novel latent replacement strategy for temporal coherence. 3) To lift these generated observations into consistent 4D representation, we propose a modulation-based refinement to mitigate inconsistencies while fully leveraging the generated information. The resulting 4D representation enables real-time, controllable rendering, marking a significant advancement in single-image-based 4D scene generation.
title Free4D: Tuning-free 4D Scene Generation with Spatial-Temporal Consistency
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2503.20785